Product feeds
DataFeedWatch agency for ecommerce brands
DataFeedWatch sits between the store catalog and every channel that wants a version of it, applying rules that rename, combine, clean and filter fields on the way out. The accounts we take over usually have one Google feed that was mapped once, with supplier titles going out untouched and disapprovals nobody has read since launch. We treat the feed as merchandising: what each product is called, which products are allowed to spend, and how quickly a price change reaches the channel.
DataFeedWatch
Product feeds · by DataFeedWatch. A text wordmark, not a logo: we are not affiliated with the vendor and make no claim to its trademarks.
What we do with it
The work inside DataFeedWatch
One source, a template per channel
The catalog is imported once, then shaped separately for each destination, because a shopping channel, a social catalog and a marketplace want different fields in different formats. Keeping one import means a fix to a product's data is made in one place and reaches every channel that carries it.
Titles built from how people search
Product titles do most of the matching work in shopping results. We build title rules by category, ordering brand, product type, key attribute and size in the sequence buyers use for that category, then keep the important words inside the length the channel displays rather than padding the field.
Identifiers, categories and required fields
Brand, condition, availability and the identifier fields decide whether a product can serve at all. We map the catalog to the channel taxonomy at a level that is specific enough to be useful, and put rules in place for the categories where the platform expects extra attributes such as size, colour, material or age group.
Filters and labels that control spend
Not every product deserves a budget. Rules exclude items that are out of stock, below margin or missing the data to compete, and custom labels group the rest by margin band, price, stock depth or seasonality so campaigns can be structured around something meaningful rather than one flat product group.
Refresh schedules and error monitoring
Update frequency is set against how quickly price and stock move in the catalog, with faster paths used for the volatile fields where the channel supports them. Disapprovals and warnings are reviewed on a schedule, and a rule change is checked against the channel's response rather than assumed to have worked.
When it fits
DataFeedWatch is the right tool when
Stores selling through more than one shopping channel
Once a catalog goes to a search channel, a social catalog and a marketplace, maintaining three separate exports by hand stops being realistic and the versions drift apart.
Catalogs the native channel app cannot express
Built-in integrations publish what the platform holds. When titles need restructuring by category, or products need excluding on margin, that logic has to live somewhere the store controls.
Merchants working with supplier or manufacturer data
Imported catalogs arrive with inconsistent titles, mixed units and gaps in the attributes channels require. Rules normalise that on the way out while the source data is corrected properly.
Teams that want campaign structure driven by the feed
Labels created in the feed become the segmentation available in the ad account, so decisions about which products get spend are made once and reflected in both places.
What to watch
Where accounts drift
Facts about defaults and costs, not criticism. These are the settings we check on every account.
Pricing follows catalog size and channel count
Plans scale with the number of products and destinations, so a large catalog with many variants should be sized before committing, including whether every variant genuinely needs to be published.
Rules become a second source of truth
After a year of edits, nobody remembers why a category is excluded. Every rule needs a note explaining its purpose, and the set needs a review each season so old exclusions do not quietly suppress products that are now worth advertising.
Changes are not instant on the channel
A rule takes effect on the next feed generation, and the channel then has to process it. Plan title or price restructuring ahead of a peak trading period rather than during it.
A feed tool cannot invent missing data
If a product has no identifier, no material and no size in the catalog, no rule creates one. The tool shortens the path to publication; the underlying data still has to be fixed in the store or the product information system.
Do we need a feed tool when the platform has its own shopping channel app?
Not always. A tidy catalog going to one channel is often fine on the native integration. The case for a feed tool starts when titles need restructuring, products need excluding on commercial rules, or a second and third channel arrive, each with different requirements.
How often should a product feed be regenerated?
Often enough that price and availability on the channel match the storefront. Stable catalogs manage on a daily schedule; stores with frequent promotions or thin stock need faster updates on those fields, using whatever supplementary or scheduled path the channel offers.
Can rules solve missing product identifiers?
They can populate the field where the value exists somewhere in the catalog, such as in a metafield or supplier column, and they can mark genuinely unbranded items correctly. What they cannot do is invent a code, so products without one are handled according to the channel's rules for that case.
Which channel do you set up first?
Whichever one already produces orders, usually the main shopping channel. Getting that feed clean lifts the campaigns already running, and it establishes the title, category and label conventions that every later channel inherits with modest adjustment.
Delivered through
Works alongside
Find the leak.
A free Growth Analysis ranks what your store should fix first, by revenue at stake.